Genomic correlates out of recombination
Pearson’s correlations along windows of 100 kb were calculated between recombination rates and the following genomic features: nucleotide diversity (?w), gene density (measured as the proportion of base pairs of the window falling into coding regions), GC content (%) and distance from the centromere to the tip of each chromosome arm (in kb). As no information exists regarding the exact position of centromeres in the Eucalyptus chromosomes, and no relationship has been yet established between the pseudochromosomes and the chromosomes in cytological observations, all chromosomes were assumed to be metacentric. Correlation significance was assessed by comparing the calculated values with those of 5000 permuted data sets that maintained the chromosomal order of all observations but that shuffled the relative positions of the two variables (Nordborg et al., 2005 ) using the function ‘sample’ in R.
Efficiency
The 2 linkage maps built with independent categories of SNPs contains 4396 SNPs having Age. grandis and you can 3991 having E. urophylla, coating similar complete recombination ranges (Fig. 1; Table S1; Cards S1). A maximum of 192 (dos.3%) nonsyntenic SNPs, was basically omitted, that’s SNPs one mapped to help you linkage teams unlike the brand new expected of these predicated on its genome standing. The newest alignment of your Age. grandis maps to the current genome type step 1.step 1 revealed set-up inconsistencies for the multiple chromosomes, way more rather into chromosomes 1, 2, 4 and you will 8 (Figs 1 anastasiadate, dos, S1). Map-centered rates regarding recombination rates have been equivalent to your several types, step 3.18 ± 1.step 1 and you will step 3.55 ± 0.8 cM Mb ?1 (Dining table S1). This new Meters arey charts (Figs dos, S1) revealed a pretty comparable slope away from recombination rate round the the chromosomes, and simply modest plateaus out of recombination was in fact seen, such as for example, on chromosomes cuatro, 9 and 10.
The amount regarding genome-broad linkage disequilibrium in the Age. grandis
Pairwise estimates of r 2 were obtained from haplotype probabilities for all pairwise distances among the Infinium SNPs on each chromosome. A total of 21 351 SNPs, an average of 1941 SNPs per chromosome, with pairwise distances varying from 135 bp up to several Mb, were used in the calculations, resulting in nearly two million pairwise estimates of r 2 per chromosome and over 21 million at the genome-wide scale for 72 sampled genomes of E. grandis. Owing to the very large number of estimates of r 2 and because the LD decay curves become an asymptote thereafter, LD decay plots are shown only up to 50 kb distances. Average genome-wide LD was r 2 = 0.131, decaying to < 0.2 within c. 5.7 kb (half-decay within 4.3 kb) while = 0.123, showing a slightly faster decay within c. 4.9 kb (half-decay within 3.7 kb) (Fig. 3). The small difference between raw and corrected r 2 is consistent with the lack of population structure between the two E. grandis provenances as indicated by the low Fst = 0.041 ± 0.06 previously calculated based on 28 658 genome-wide SNPs (Silva-ong the c. 4000 linkage mapped markers in E. grandis and r 2 plotted against the distance in cM. With map resolution of c. 0.3 cM, corresponding to c. 106 kb, no r 2 estimate was larger than 0.2, consistent with the decay observed at c. 4–6 kb (Fig. S2). No impact of the few assembly inconsistencies of the E. grandis genome version 1.1 was seen on the pattern of LD decay and no difference was observed when rarer SNPs with MAF > 0.01 were included in the estimation of r 2 (Fig. S3).
Population-scaled recombination rates in E. grandis
About three remedies for receive genome-wide rates regarding ? have been undertaken using two independent experimental study establishes related almost thirteen mil SNPs about pooled sequence analysis as well as 21 100 Infinium SNPs. Chromosome-certain prices off ? was acquired for different genomic screen versions in terms of SNP amounts that have LDH from the and you may H otspotter (Dining table step one). Estimates was indeed determined across every chromosomes within the 2402 overlapping bins of 15 SNPs (indicate container size, 250 kb; SNP density, 1/sixteen.7 kb) as well as in 242 overlapping pots regarding 105 SNPs (suggest bin size, 2500 kb; SNP density, 1/ kb). Absolutely nothing type was viewed between the two prices and you may across the chromosomes. Prices gotten by the H otspotter had been generally speaking doubly highest just like the those individuals from LDH on , probably highlighting different presumptions concerning your LD brand of the fresh one or two estimation tips. Thus, the brand new genome-greater rates away from productive populace items acquired because of the equating ? so you’re able to brand new recombination speed c was indeed including twice as high which have H otspotter . The latest genome-broad rates away from ? gotten that have Infinium SNPs converged to truly similar viewpoints and you can magnitudes to the guess acquired to your pooled sequencing research using mlRho and to quotes based on population-top LD (Table dos). Even though some level of testing bias inherent into the strategy appears to-be establish when comparing brand new quotes from ? out-of H otspotter and you can mlRho and also involving the rates off LDH at and you will inhabitants-peak LD, the values failed to differ by the > 30%. Nonetheless, the second a couple estimates are two to three minutes smaller compared to the initial a few. All the estimates out of ? features higher standard deviations, reflecting the latest expected genome-large variation in the recombination, and especially thus on estimate regarding sequencing research, possibly due to the much larger number of nucleotides examined.